COMPARISON BETWEEN THE TARIM AND OTHER CENTRAL ASIAN BASINS. 375
kamisch-thickets, but connected together by channels, precisely as in the existing lake of Lop-nor. It is to this period that we must assign the upper part of the deposits on the terraces of the Mogutun-gol. At a later date these small lakes disappeared, and the Bulungir began to cut its way through the former lacustrine deposits; though when it was in high flood, the river was at the same time able to overflow the salt expanses and meadow-lands nearest to its banks. This is the epoch to which Mr Grum-Grschimajlo refers, when he speaks of the artificial irrigation of the meadows by the Mongols during the Juan period, that is about the year 1368 A. D. At the present time it would only be possible to irrigate in this way the eastern part of the depression, around the well of Si-dun; in its western part the stream has cut too deeply into the former bed of the lake to permit of the water flowing over.»
»With regard to the lake of Alak-tschi, Mr Grum-Grschimajlo says in the beginning of the passage quoted, 'In the same way the lake Alak-tschi, as we shall see lower down, can hardly have existed, at all events within historic times' ; and on the next page he adds, that this lake is shown on the maps of D'Anville, Grimm, and Jakinf, and that it actually does exist.»
»It will therefore be clear which of us is right, and I ask Mr Grum-Grschimajlo, what profit it will serve to continue the controversy? Would it not be better to adduce historical and cartographical corrections, in support of the former existence of these lakes, and drop the hasty phrase made use of in his book, that these lakes could not exist. But in the beginning of the passage I have again quoted from him he once more repeats the same phrase, although in a less aggressive form; then he reproduces my first refutation incorrectly; and finally takes pains to prove that it is unfounded. In doing that he is however most unfortunate in his choice of method; for it is evident from his own words, that one of these lakes was formed within historical times by artificial means, and that the other exists even at the present day, and there really is nothing in the topographical conditions to prevent either the one or the other lake from coming into existence.»
For so well-trained a geologist and geographer as Obrutscheff it was not difficult to interpret the relief features of the country, and to trace out their connection with the hydrographical relations that obtain in that region; and his rigidly critical statement of the problem is a guarantee that he has not made a mistake. The similarity between this problem and the Lop-nor problem, to which I alluded in the beginning of this chapter, lies in the fact that both lacustrine groups belong to the same gigantic latitudinal valley, and that the solution offered in both cases alike has given rise to a lively controversy. Grum-Grschimajlo has employed all his acumen and learning in defence of the opinion to which he has given utterance; Kosloff has done his very best to maintain Prschevalskij's right to be the discoverer of the Lop-nor. But in both cases alike these laudable efforts have had to yield before exact investigations into the actual facts, and scientific truth, to which all other considerations must yield, has gained the victory.
But from the physico-geographical point of view there exists great dissimilarity between the two lakes. The Lop-nor is an ambulatory lake, a terminal basin; the Chua-chai-tsi occupied a permanent situation, and was not a terminal lake, but one
II e di n, yourney in Central Asia. II. 48
IIIFカラー高解像度
